Professional Documents
Culture Documents
Probability Terminology
Random variable: uncertain number Outcome: realization of random variable Event: set of one or more outcomes Mutually exclusive: cannot both happen Exhaustive: set of events includes all possible outcomes
Types of Probability
Empirical: based on analysis of data Subjective: based on personal perception A priori: based on reasoning, not experience
Joint Probability
The probability that both of two events will occur is their joint probability
Example using conditional probability: P (interest rates will increase) = P(I) = 40% P (recession given a rate increase) = P(R|I) = 70%
Probability of a recession and an increase in rates, P(RI) = P(R|I) P(I) = 0.7 0.4 = 28%
An Investment Tree
Prob of good stock performance 30% Prob of good economy Expected EPS = $1.51 Prob of poor economy 60% 40% EPS = $1.80 Prob = 18% EPS = $1.70 Prob = 42%
70%
60%
40%
Covariance
Covariance: A measure of how two variables move together Values range from minus infinity to positive infinity Units of covariance difficult to interpret Covariance positive when the two variables tend to be above (below) their expected values at the same time
For each observation, multiply each probability times the product of the two random variables deviations from their means and sum them
Correlation
Correlation: A standardized measure of the linear
Correlation
Example: The covariance between two assets is 0.0046, A = 0.0623 and B = 0.0991. What is the correlation between the two assets (A,B)?
the variance
E(RB)=18%
RA= 20%
Probabilities
0.60
0.25
RA= 15%
RA= 4% E(RA)=13%
CovAB=
Bayes Formula
Prob. of interest rate cut (C) 60% 70% 30% 20% 40% Prob. of no interest rate cut
Prob (C|G) = 42/(42 + 8) = 84%
Prob (C|G) = [Prob(G|C) Prob(C)]/Prob(G) Prob (C/G) = (70% * 60% )/ (42% +8%)
80%
10! 2,520 5! 3! 2!
possible labels, we can use the combination formula (binomial formula) Example: You have 5 stocks and want to place orders to sell 3 of them. How many different combinations of 3 stocks are there?
permutation formula:
You have 5 stocks and want to sell 3, one at a
time. The order of the stock sales matters. How many ways are there to choose the 3 stocks to sell in order?
matter?
5 2nd nPr 3 = 60
Probability Functions
A probability function, p(x), gives the probability
that a discrete random variable will take on the value x e.g. p(x) = x/15 for X = {1,2,3,4,5} p(3) = 20% A probability density function (pdf), f(x) can be used to evaluate the probability that a continuous random variable with take on a value within a range A cumulative distribution function (cdf), F(x), gives the probability that a random variable will be less than or equal to a given value
variables is also normally distributed Probabilities decrease further from the mean, but the tails go on forever Multivariate normal: more than one r.v., correlation between their outcomes
distributed) on a portfolio over many years is 11%, and the standard deviation of returns is 8%. A 95% confidence interval on next years return is 11% + (1.96)(8%) = 4.7% to 26.7%
so that mean = 0 and standard deviation = 1 To standardize a random variable, calculate the z-value Subtract the mean (so mean = 0) and divide by standard deviation (so = 1)
the table this is 0.5793 Since the distribution is symmetric, for negative values we take 1 minus the table value Probability of values less than $3.70 is 1 0.5793 = 42.07% With a z-table for negatives, F(-0.20) = 0.4207